Sleep Deprivation and Emotion Recognition

IN THIS ISSUE OF SLEEP, VAN DER HELM AND COLLEAGUES¹ DESCRIBE FOR THE FIRST TIME THE IMPACT OF ONE NIGHT OF SLEEP DEPRIVATION ON THE ability to recognize human facial emotion intensities. The experimental design included an emotional face recognition task using a gradiently morphed version of the Ekman Pictures of facial affect set for the emotions ‘Sad', ‘Happy' and ‘Angry', ranging from neutral to increasingly emotional. Healthy young adults were randomized to a sleep deprivation or a sleep rested control condition, with tests being performed thrice in the sleep deprivation condition (after 30h of behavioral sleep deprivation, after recovery sleep, and three weeks after initial tests) and twice in the control group (two consecutive days after sleep rest). The authors observed a marked impairment of the accurate recognition of subtle emotion intensities after a single night of sleep deprivation. These deficits were especially evident for recognition of ‘Happy' and ‘Angry' expressions, and were particularly pronounced in women. The impairments in emotion recognition were improved after a subsequent night of recovery sleep.

Accurate emotional face judgments rely on the human ability to be empathetic, literally to “feel into” another person (after the German word “einfuehlen”) and are crucial for reciprocal affective communication. Abnormal face processing has been well studied in developmental disorders such as autism, and is consistently associated with social dysfunction². Although impairment of human facial emotion recognition after sleep deprivation is rather functional,¹ it may have implications for social interactions at a societal level, given the putative decrease in sleep duration over the past decades.^3,4 Deficient emotion recognition may be even more relevant on an interpersonal level for specific populations that frequently experience acute or chronic sleep deprivation: medical personnel, shift workers or parents of young children, to name only a few. While the findings of van der Helm and colleagues¹ presented in this issue of SLEEP are preliminary due to study limitations, the authors pave the way for a new line of research investigating the effects of sleep curtailment on human emotion recognition.

The described deficit in identifying emotion intensity after one night of sleep deprivation was found to be particularly pronounced in women, suggesting a gender specific vulnerability to the effects of sleep deprivation. As emotion recognition is related to subjective mood states,⁵ these findings may be associated with observed gender differences in mood regulation after sleep curtailment. Subjective mood changes by altering the duration and timing of sleep episodes.⁶ Indeed, our group recently demonstrated that women displayed a significantly greater deterioration of well-being under sleep deprivation compared to men,⁷ in a constant routine protocol assessing the impact of homeostatic and circadian factors on well-being under highly controlled conditions. The impairment of well-being under sleep deprivation in young women was associated with a rise in salivary cortisol levels,⁷ which may reflect higher levels of stress experienced by women under sleep deprivation. Future studies need to address the interrelation between subjective mood states, self-reported and objective stress and sleep loss on emotion recognition.

In contrast, van der Helm et al.¹ did not observe significant effects of sleep deprivation on emotion recognition in men, accounted for by a higher variance in test results in male participants after sleep deprivation.¹ Circadian modulation contributes substantially to a multitude of physiological and behavioral parameters, such as mood, sleepiness and neurobehavioral performance.^8,9 A specific clock time for testing (4pm in this study) may correspond to different circadian times in individuals. Given that delayed sleep phases are highly common in adolescence and young adulthood,¹⁰ and may be more pronounced in young men, differences in circadian time may have represented a significant bias in terms of the gender differences observed in this study, possibly accounting for some of the increased variance in male subjects. Future studies may adjust their study design in accordance to individual chronotypes (i.e. individual traits of diurnal preference) thus scheduling test times in accordance to each subject's circadian phase.

Interestingly, the impairment of emotion recognition observed in the present study emerged in the mid-spectrum range, in response to emotionally ambiguous facial expressions, whereas the recognition of either neutral or the most emotional facial expressions was unaffected by one night of sleep deprivation. These results suggest that the basic aspects of perception of facial emotions may still be intact, while the threshold for emotion recognition may be modified: the intensity of facial emotion required to activate the respective networks supporting each of the emotions (here, ‘Angry' and ‘Happy') may be greater under conditions of sleep deprivation.

Finally, it is important to note that subjective sleepiness, as measured by the Stanford Sleepiness Scale, did not affect the ability to accurately identify emotion intensities. Though subjects were overall sleepier after sleep deprivation, subjective sleepiness scores were not significantly correlated with mean emotion ratings for any of the three tested emotions.

The study of van der Helm et al.¹ points to important deficiencies in our knowledge of the relation between sleep loss and emotion recognition. What are the effects of chronic partial sleep deprivation on emotion recognition in a society where suboptimal sleep duration is increasingly the rule? How do circadian factors affect our ability to accurately identify human facial emotion recognition, alone or in interaction with homeostatic factors? What are the implications of the present findings for patients suffering from sleep disorders, mood disorders or both? How do subjective mood state and perceived as well as objective stress interfere with human emotion recognition? Are there any age effects modulating the effects of sleep deprivation on human facial emotion recognition? Overall, a better understanding of the factors that determine individual differences in vulnerability to sleep deprivation would be useful in interpreting these data. Finally, future studies should employ objective measures of sleep using polysomnography and measures of the sleep-wake-cycle with actigraphy in order to verify protocol compliance, and assess sleep and sleep-wake related parameters that sustain emotion recognition.

DISCLOSURE STATEMENT

Dr. Schroder has indicated no financial conflicts of interest.